NK cells have potent antitumor capacity. They are enriched in the human liver, with a large subset specialized for tissue-residence. The potential for liver-resident versus liver-infiltrating NK cells to populate, and exert antitumor functions in, human liver tumors has not been studied. We examined liver-resident and liver-infiltrating NK cells directly ex vivo from human hepatocellular carcinomas (HCCs) and liver colorectal (CRC) metastases, compared with matched uninvolved liver tissue. We found that NK cells were highly prevalent in both HCC and liver CRC metastases, although at lower frequencies than unaffected liver. Up to 79% of intratumoral NK cells had the CXCR6+CD69+ liver-resident phenotype. Direct ex vivo staining showed that liver-resident NK cells had increased NKG2D expression compared to their non-resident counterparts, but both subsets had NKG2D downregulation within liver tumors compared to uninvolved liver. Proliferation of intratumoral NK cells (identified by Ki67) was selectively impaired in those with the most marked NKG2D downregulation. Human liver tumor NK cells were functionally impaired, with reduced capacity for cytotoxicity and production of cytokines, even when compared to the hypo-functional tissue-resident NK cells in unaffected liver. Coculture of human liver NK cells with the human hepatoma cell line PLC/PRF/5, or with autologous HCC, recapitulated the defects observed in NK cells extracted from tumors, with downmodulation of NKG2D, cytokine production, and target cell cytotoxicity. Transwells and conditioned media confirmed a requirement for cell contact with PLC/PRF/5 to impose NK cell inhibition. IL-15 was able to recover antitumor functionality in NK cells inhibited by in vitro exposure to HCC cell lines or extracted directly from HCC. In summary, our data suggest that the impaired antitumor function of local NK cells reflects a combination of the tolerogenic features inherent to liver-resident NK cells together with additional contact-dependent inhibition imposed by HCC itself. The demonstration that IL-15 can recover hepatic NK cell function following tumor exposure supports its inclusion in immunotherapy strategies.

BackgroundEnvironmental enteric dysfunction (EED) is an acquired syndrome of impaired gastrointestinal mucosal barrier function that is thought to play a key role in the pathogenesis of stunting in early life. It has been conceptualized as an adaptive response to excess environmental pathogen exposure. However, it is clinically similar to other inflammatory enteropathies, which result from both host and environmental triggers, and for which immunomodulation is a cornerstone of therapy.MethodsIn this pilot double-blind randomized placebo-controlled trial, 44 children with severe acute malnutrition and evidence of EED were assigned to treatment with mesalazine or placebo for 28 days during nutritional rehabilitation. Primary outcomes were safety and acceptability of the intervention.ResultsTreatment with mesalazine was safe: there was no excess of adverse events, evidence of deterioration in intestinal barrier integrity or impact on nutritional recovery. There were modest reductions in several inflammatory markers with mesalazine compared to placebo. Depression of the growth hormone – insulin-like growth factor-1 axis was evident at enrollment and associated with inflammatory activation. Increases in the former and decreases in the latter correlated with linear growth.ConclusionsIntestinal inflammation in EED is non-essential for mucosal homeostasis and is at least partly maladaptive. Further trials of gut-specific immunomodulatory therapies targeting host inflammatory activation in order to optimize the growth benefits of nutritional rehabilitation and to address stunting are warranted. Funded by The Wellcome Trust.Trial registrationRegistered at Clinicaltrials.gov NCT01841099.Electronic supplementary materialThe online version of this article (doi:10.1186/s12916-014-0133-2) contains supplementary material, which is available to authorized users.

BackgroundReady-to-use therapeutic foods (RUTF) are lipid-based pastes widely used in the treatment of acute malnutrition. Current specifications for RUTF permit a high n-6 polyunsaturated fatty acid (PUFA) content and low n-3 PUFA, with no stipulated requirements for preformed long-chain n-3 PUFA. The objective of this study was to develop an RUTF with elevated short-chain n-3 PUFA and measure its impact, with and without fish oil supplementation, on children’s PUFA status during treatment of severe acute malnutrition.MethodsThis randomized controlled trial in children with severe acute malnutrition in rural Kenya included 60 children aged 6 to 50 months who were randomized to receive i) RUTF with standard composition; ii) RUTF with elevated short chain n-3 PUFA; or iii) RUTF with elevated short chain n-3 PUFA plus fish oil capsules. Participants were followed-up for 3 months. The primary outcome was erythrocyte PUFA composition.ResultsErythrocyte docosahexaenoic acid (DHA) content declined from baseline in the two arms not receiving fish oil. Erythrocyte long-chain n-3 PUFA content following treatment was significantly higher for participants in the arm receiving fish oil than for those in the arms receiving RUTF with elevated short chain n-3 PUFA or standard RUTF alone: 3 months after enrolment, DHA content was 6.3% (interquartile range 6.0–7.3), 4.5% (3.9–4.9), and 3.9% (2.4–5.7) of total erythrocyte fatty acids (P <0.001), respectively, while eicosapentaenoic acid (EPA) content was 2.0% (1.5–2.6), 0.7% (0.6–0.8), and 0.4% (0.3–0.5) (P <0.001). RUTF with elevated short chain n-3 PUFA and fish oil capsules were acceptable to participants and carers, and there were no significant differences in safety outcomes.ConclusionsPUFA requirements of children with SAM are not met by current formulations of RUTF, or by an RUTF with elevated short-chain n-3 PUFA without additional preformed long-chain n-3 PUFA. Clinical and growth implications of revised formulations need to be addressed in large clinical trials.Trial registrationClinicaltrials.gov NCT01593969. Registered 4 May 2012.Electronic supplementary materialThe online version of this article (doi:10.1186/s12916-015-0315-6) contains supplementary material, which is available to authorized users.

Implementation of successful prevention of mother-to-child transmission of HIV strategies has resulted in an increased population of HIV-exposed uninfected (HEU) infants. HEU infants have higher rates of morbidity and mortality than HIV-unexposed (HU) infants. Numerous factors may contribute to poor health in HEU infants including immunological alterations. The present study assessed T-cell phenotype and function in HEU infants with a focus on memory Th1 responses to vaccination. We compared cross-sectionally selected parameters at 3 and 12 months of age in HIV-exposed (n = 42) and HU (n = 28) Kenyan infants. We measured ex vivo activated and bulk memory CD4 and CD8 T-cells and regulatory T-cells by flow cytometry. In addition, we measured the magnitude, quality and memory phenotype of antigen-specific T-cell responses to Bacillus Calmette-Guerin and Tetanus Toxoid vaccine antigens, and the magnitude and quality of the T cell response following polyclonal stimulation with staphylococcal enterotoxin B. Finally, the influence of maternal disease markers on the immunological parameters measured was assessed in HEU infants. Few perturbations were detected in ex vivo T-cell subsets, though amongst HEU infants maternal HIV viral load positively correlated with CD8 T cell immune activation at 12 months. Conversely, we observed age-dependent differences in the magnitude and polyfunctionality of IL-2 and TNF-α responses to vaccine antigens particularly in Th1 cells. These changes mirrored those seen following polyclonal stimulation, where at 3 months, cytokine responses were higher in HEU infants compared to HU infants, and at 12 months, HEU infant cytokine responses were consistently lower than those seen in HU infants. Finally, reduced effector memory Th1 responses to vaccine antigens were observed in HEU infants at 3 and 12 months and higher central memory Th1 responses to M. tuberculosis antigens were observed at 3 months only. Long-term monitoring of vaccine efficacy and T-cell immunity in this vulnerable population is warranted.

Controlled human malaria challenge studies in semi-immune volunteers provide an unparalleled opportunity to robustly identify mechanistic correlates of protection. We leveraged this platform to undertake a head-to-head comparison of seven functional antibody assays that are relevant to immunity against the erythrocytic merozoite stage of Plasmodium falciparum. Fc-mediated effector functions were strongly associated with protection from clinical symptoms of malaria and exponential parasite multiplication while the gold standard growth inhibition assay was not. The breadth of Fc-mediated effector function clearly discriminated grades of clinical immunity following challenge. These findings present a paradigm shift in the understanding of the mechanisms that underpin immunity to malaria and have important implications for vaccine development.